Subject: Re: Guidance Needed for PostgreSQL Upgrade from 12 to 15
To: Scott Ribe <scott_ribe@elevated-dev.com>
Cc: pgsql-admin@lists.postgresql.org
References: 
 <CAL5GnisA0RJ=Rd5e6yw4nsCB+DtAS1=iquBKTH76MknnTsHrPg@mail.gmail.com>
 <CANzqJaA1VyGqPUQaeT9vfxNV1C3rQJSdMHa9xtzx+rJQhkY2gw@mail.gmail.com>
 <CAL5Gnitdyw7vCp4-0whmmJ-RC7Z9yQVx=xrtfQO6-Lxcfvqrqw@mail.gmail.com>
 <CANzqJaAL-vn0iwP8QGEcVn9RqLv9Uw12qG6fV0h5+AkwW1oGBQ@mail.gmail.com>
 <0b8c1b80-93f5-40a7-b67b-19e58207c12c@cloud.gatewaynet.com>
 <9724B260-8B0E-4B03-A66A-784F792459C9@elevated-dev.com>
 <355729f6-74a0-1b5c-a065-bd919507a490@vianet.ca>
 <39901423-2003-493F-8A5C-12D1B1062F00@elevated-dev.com>
From: Kris Deugau <kdeugau@vianet.ca>
Organization: Vianet Internet Solutions
Message-ID: <46c1255a-725f-3f0d-a18f-e1a9c20dd36e@vianet.ca>
Date: Tue, 19 Nov 2024 15:40:51 -0500
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 SeaMonkey/2.53.19
MIME-Version: 1.0
In-Reply-To: <39901423-2003-493F-8A5C-12D1B1062F00@elevated-dev.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
 <https://www.postgresql.org/message-id/46c1255a-725f-3f0d-a18f-e1a9c20dd36e%40vianet.ca>
Precedence: bulk

Scott Ribe wrote:
>> On Nov 19, 2024, at 10:32 AM, Kris Deugau <kdeugau@vianet.ca> wrote:
>>
>> as otherwise the old PG won't start properly, due to the changes made by pg_upgrade.
> 
> Not true, pg_upgrade leaves it in a state where either can be started. By taking the snapshot after, if you roll back to it, you can attempt changes on either the old or new. (Taking care to take additional snapshots as needed, to preserve the ability to roll back to this state.)

I stand corrected.  I hadn't read the docs on pg_upgrade for quite a 
while, but after reading the last section in 
https://www.postgresql.org/docs/current/pgupgrade.html:

"If you did not start the new cluster, the old cluster was unmodified 
except that, when linking started, a .old suffix was appended to 
$PGDATA/global/pg_control. To reuse the old cluster, remove the .old 
suffix from $PGDATA/global/pg_control; you can then restart the old 
cluster."

I see what you mean.

-kgd